Haven't had time to do it yet, but aside from keyword filtering I think a good solution is to block replies posted within the same second as the original post. A real human couldn't reply within 1 second. Yes they can just add a sleep into the script, but it adds a layer to the spam filter. At the intersection of all the mitigations, spam will be reduced.
nostr:note1x3w56pr8evc2mhlssjxu8dwvjahnsue2v3605awr0xcustckql5q39ul9u
if the people you follow have marked the post as spam, odds increase it is and you can down prioritize it #peercuration
But that would be mean towards nostr:npub1c4vv0nrfh0drcfchs2mndw4uvjkd9k393v20x4kme2txev9hhz0qw4cqk7!
Nah, the bots will just adjust around the limit and delay their replies a few more seconds. Web of trust relays are the only way free relays will survive the spam I suspect.
I’m all for Web of Trust relays but isn’t all security a cat-and-mouse game? Let’s at least force the bot to make a change in the short term. Web of Trust relays feel like a longer-term, harder-to-implement solution.
Probably even a *better* longer-term, harder-to-implement solution. But let’s make the quick fix now.
I muted “, from wss://“. May catch some legitimate posts, but worth it.
The bot’s whole schtick is the relay ad, so that’s my target.
I agree. Even if they make the script sleep, this feels like a good short-term, quick solution.
1 second is nothing. How dare you underestimate my reply guy skills.
